About three to five percent — and in several careful studies, nothing at all#
Does your metabolism stay low after weight loss? Measured in people who have already lost the weight and are eating at maintenance, the shortfall is real but small. Pooling the published comparisons of formerly obese adults against never-obese controls, resting metabolic rate adjusted for fat-free and fat mass came out 5.1 percent lower (95% CI 1.7 to 8.6) in the study-level analysis of 12 studies, and 2.9 percent lower in the pooled individual data from 124 formerly obese and 121 control subjects — where it did not reach significance (p = 0.09). The authors' summary was a 3 to 5 percent lower mean relative resting rate1.
On a resting rate near 1,500 calories, five percent is about 75 calories a day (our arithmetic on their percentage, not a figure the paper publishes). That is a genuine headwind and it is roughly a seventh of the number that circulates online, which traces to a single extraordinary cohort. What follows is the evidence from ordinary weight-reduced people — the ones who lost 15 or 30 kilos rather than 58 — where the picture is smaller, partly reversible, and still genuinely contested. The mechanism of the adaptation and why researchers cannot agree how to measure it belongs to adaptive thermogenesis explained; the claim that a diet can shut your metabolism down is dealt with in is 'starvation mode' real. This page asks the narrower thing people actually mean: once it is over, am I permanently cheaper to run?
One caution belongs on Astrup's number before anyone spends it. His team pointed out that the average difference "could be explained by a low RMR being more frequent among the formerly obese subjects"1 — that a low resting rate may be commoner in people who became obese in the first place, rather than being something the dieting did. A cross-sectional comparison cannot separate those two stories, and most of the internet's version assumes the second without argument.
People who have actually kept it off look close to normal#
The sharpest test is to find people who have already done the hard part and put them on a metabolic cart.
Forty adults from the National Weight Control Registry — each having lost at least 13.6 kg and held it for over a year — were measured against 46 weight-matched controls. Once lean mass, fat mass, age and sex were accounted for, resting rate came out at 5,926 ± 106 kJ/day in the reduced-obese group and 6,015 ± 104 in the controls (P = 0.35). About 21 calories apart, on our conversion of their kilojoule figures. The authors' reading: "in at least some reduced-obese individuals there does not seem to be a permanent obligatory reduction in RMR beyond the expected reduction for a reduced lean mass"2. One asymmetry did appear — the maintainers' fasting respiratory quotient was slightly higher (0.807 vs 0.791, P = 0.05), consistent with oxidising proportionally less fat at rest. That is a claim about fuel mix, not about how much energy is being spent, and it is routinely mis-cited as the latter.
A modern replication with tighter controls landed in the same place. Thirty-four long-term maintainers were compared against 35 normal-weight and 33 overweight or obese controls; their measured resting expenditure was predicted to within about 1 percent by equations built from the control groups, and within 3 percent by standard equations. But individual differences between measured and predicted ran from −257 to +163 kcal/day3.
The average maintainer's resting metabolism is unremarkable. The spread around that average is four hundred calories wide, which is why the average describes almost nobody in it.
One disclosure belongs on the page, applied in the direction that is inconvenient for this article's own conclusion. Two authors of that 2018 paper report stock options in weight-management companies and one holds an advisory-board role with a commercial diet brand, and the same research group produced the 1999 registry study above. "Your metabolism is not permanently broken" is a commercially useful finding for a weight-management business, and the two most reassuring datasets here come out of one lab. Indirect calorimetry is hard to spin, so this is not a reason to discard the measurements. It is a reason the next section matters.
The study that finds it persists — and whose own numbers show it partly lifting#
The strongest counterweight comes from a different design with no such ties. Ninety-one adults (22 men, 69 women, mean BMI 31.9) completed an eight-week very-low-energy diet followed by 44 weeks of weight maintenance. The ratio of measured to predicted resting rate fell from 1.004 ± 0.077 at baseline to 0.963 ± 0.073 immediately after the diet, then read 0.983 ± 0.063 at 20 weeks and 0.984 ± 0.068 at 52 weeks. The conclusion is explicit: "weight loss results in adaptive thermogenesis, and there is no indication for a change in adaptive thermogenesis up to 1 y, when weight loss is maintained"4.
That conclusion is about the maintenance year, and it is correct: from week 20 to week 52 the ratio does not move. But look at what happened just before the maintenance year began. The diet drove the ratio down by 0.041, and by week 20 it had come back up by 0.020 — roughly half the fall, recovered (our arithmetic on their published ratios). What persisted was the other half: about 1.6 percent below prediction, holding steady for the rest of the year. Both readings of that dataset are true, and the popular one quotes only the second.
So one line of work finds no detectable gap and another finds one that lasts a year. What separates them is not the physiology. It is the denominator.
| Wyatt, Ostendorf | Camps | |
|---|---|---|
| Compared against | Weight-matched or never-obese controls | Each participant's own baseline prediction |
| Question answered | Do you burn less than someone who has always been this size? | Do you burn less than you did, for your size? |
| Result | No statistically detectable difference | ~1.6% below prediction, sustained to 52 weeks |
Now set the numbers beside each other instead of the conclusions. Camps's persistent gap is 1.6 percent. Astrup's meta-analysis put weight-stable formerly obese adults 2.9 to 5.1 percent below controls. These are four estimates clustered between roughly one and five percent, arguing about whether that band is a finding or a rounding error. Nobody in the dispute is measuring an ordinary maintainer running several hundred calories below prediction. (A separate inpatient study that measured the same women in and out of an active deficit found the suppression largely disappeared once they were weight-stable — that comparison anchors what your maintenance calories actually become.)
The variable that reconciles all of it#
The 2018 study contains one result that does more explanatory work than anything else in this literature. Across its maintainers, the shortfall between measured and predicted resting expenditure correlated with how much weight had been lost (r = 0.36, P < 0.05) and not at all with how long it had been kept off (r = 0.04, P = 0.81)3.
That single pair of correlations settles the argument the internet keeps having. The famous several-hundred-calorie suppression comes from television competitors who lost an average of 58 kg — the far end of the loss axis, and the case handled in is 'starvation mode' real. Registry maintainers who lost 15 to 30 kg sit low on that same axis and show correspondingly little. The relationship is with magnitude, not with time, and that carries two consequences at once: a very large loss does buy a real metabolic cost, and the cost does not deepen the longer you hold the new weight. Nothing is quietly getting worse while you maintain.
It also explains why the two literatures look irreconcilable when they are quoted side by side. They are samples drawn from opposite ends of a variable that nobody names in the headline.
Planning around seventy calories#
If the honest central estimate is a few percent — call it 50 to 90 calories a day after a substantial loss, less after a modest one — the practical consequences are unglamorous and mostly not about metabolism at all.
- Budget for it once and stop thinking about it. Seventy calories is smaller than the error in almost any calorie estimate you will make today. It is not a reason to recalculate weekly; it is a reason to set maintenance from your own weight trend rather than a formula.
- Expect no further decline. The shortfall tracks how much you lost, not how long you have maintained. Year three is not metabolically harder than year one.
- Look at appetite before you look at metabolism. The reason maintenance is hard sits mostly on the intake side, and it is several times larger than the expenditure side — the behaviours that hold a loss are in how to keep the weight off, and the profile of people who manage it in what maintainers actually do.
- Defend lean mass and daily movement, because those set the resting rate that dominates your total — the organ-by-organ version is in metabolism explained.
The fear underneath this question is that dieting leaves permanent damage. The measurements do not support that. What they support is milder and stranger: after a large loss you are a slightly cheaper body to run than someone who has always been your size, by an amount that is hard to detect, does not grow with time, and is comfortably outweighed by the hunger that arrives with it.
FAQ#
Is my metabolism damaged from years of dieting?#
Not in any sense the measurements can find. Across the pooled comparisons, weight-stable formerly obese adults run about 3 to 5 percent below matched never-obese controls, and in the individual-subject analysis of 124 formerly obese and 121 control subjects that difference did not reach statistical significance. The same authors noted their result could partly reflect a low resting rate being commoner among people who became obese, rather than an effect of the dieting itself.
Do I burn less than someone who has always been my current weight?#
Slightly, or possibly not at all — the studies split and the difference is small either way. Registry members holding a loss of at least 13.6 kg for over a year had an adjusted resting rate of 5,926 kJ/day against 6,015 in weight-matched controls, a gap of roughly 21 calories that was not statistically significant. Individual variation dwarfs the group difference: across 34 long-term maintainers, measured minus predicted resting expenditure ran from 257 calories below prediction to 163 above.
Does the slowdown get worse the longer you keep the weight off?#
No, and it is one of the more reassuring findings in the field. In long-term maintainers the shortfall between measured and predicted resting expenditure correlated with how much weight had been lost (r = 0.36) and showed essentially no relationship with how long the loss had been held (r = 0.04, P = 0.81). Whatever metabolic cost a large loss carries appears to be set by the size of the loss, not accumulated over the years you spend maintaining it.
Sources#
- Astrup A, Gøtzsche PC, van de Werken K, et al. Meta-analysis of resting metabolic rate in formerly obese subjects. Am J Clin Nutr. 1999;69(6):1117-1122.
- Wyatt HR, Grunwald GK, Seagle HM, et al. Resting energy expenditure in reduced-obese subjects in the National Weight Control Registry. Am J Clin Nutr. 1999;69(6):1189-1193.
- Ostendorf DM, Melanson EL, Caldwell AE, et al. No consistent evidence of a disproportionately low resting energy expenditure in long-term successful weight-loss maintainers. Am J Clin Nutr. 2018;108(4):658-666.
- Camps SG, Verhoef SP, Westerterp KR. Weight loss, weight maintenance, and adaptive thermogenesis. Am J Clin Nutr. 2013;97(5):990-994.
